How to Run a Broken Link and Error Audit

A broken link is any hyperlink that does not resolve to a live, accessible page. From a search engine's perspective, broken links create three distinct problems: they waste crawl budget on dead endpoints, they break the flow of PageRank between pages, and they signal poor content quality to ranking algorithms. From a user perspective, they erode trust — a visitor who clicks a link and lands on a 404 is more likely to leave the site entirely.

Crawl budget and PageRank leakage

Googlebot has a fixed crawl budget for each domain — a limit on how many pages it will crawl within a given timeframe. Every request to a broken URL consumes that budget and returns nothing useful. On large sites with thousands of pages, broken links can consume a meaningful fraction of crawl allocation, leaving newly published or updated pages undiscovered for weeks. The second issue — PageRank leakage — is more nuanced: link equity flows out through links but cannot flow into a 404 page, so it is effectively lost rather than redistributed.

The compounding effect of site migrations

Most broken links are not introduced gradually — they arrive in batches during URL restructures, CMS migrations, domain changes, and page deletions. A site that moves from `/blog/post-title` to `/articles/post-title` without setting up 301 redirects will break every internal and external link to every affected URL simultaneously. Running a complete audit immediately after any structural change catches the damage before crawlers index the broken state.

• Crawl budget waste: Every 404 response consumes crawl quota with no indexing benefit.
• PageRank leakage: Link equity sent to a dead URL is lost rather than redistributed across the site.
• Quality signals: Pages with high broken link density are associated with lower-quality content in ranking models.
• User experience: A visitor who clicks a dead link is significantly more likely to abandon the session.
• Broken anchor text signals: Internal link anchor text helps Google understand page topics — dead links remove that signal.

A complete broken link and error audit covers five distinct problem categories. Each category requires a different tool and produces different types of fixes. Understanding the full scope before you start prevents the common mistake of treating a broken link audit as simply "find dead URLs" — there are four other error types that are equally damaging and often missed.

The five error categories

• Broken internal links: Links between pages on your own site that point to deleted, renamed, or moved pages, including broken anchor fragment links (`#section-id`).
• Broken external links: Outbound links to third-party URLs that return errors, are malformed, use HTTP instead of HTTPS, or point to domains that have expired.
• Redirect chains and loops: URLs that redirect through multiple intermediate hops (chain) or that redirect in a cycle (loop), wasting crawl budget and degrading link equity.
• Canonical conflicts: Pages with canonical tags pointing to the wrong URL, using relative paths, referencing non-HTTPS targets, or self-referencing incorrectly.
• Sitemap and robots.txt errors: Sitemaps that include noindex pages, malformed XML, invalid URLs, or missing required elements; robots.txt files with directives that block key pages.

A broken link audit is not a one-time task — it is a recurring quality check that should run after every significant content change and every deployment.

Internal broken links fall into two categories: broken page links (the destination URL returns a 404 or has been removed from the site) and broken fragment links (the page loads correctly but the `#anchor-id` does not exist on the page). The second type is consistently underdetected because standard crawlers only check whether a page loads — they do not verify whether the specified fragment target exists within the loaded HTML.

External broken links point to third-party URLs you do not control. The most common causes are websites going offline, pages being deleted, URLs changing structure without preserving the old path, and domains expiring. From an SEO perspective, external broken links are a quality signal — a page full of dead outbound references looks stale and unreliable. From a user experience perspective, they are simply frustrating.

What the Broken External Link Checker detects

Paste your page HTML into the Broken External Link Checker and the tool analyses every outbound `href` attribute that points to an external domain. It flags malformed URLs (missing protocol, invalid characters, improperly encoded path segments), insecure HTTP links on HTTPS pages, and suspicious link patterns associated with low-quality outbound link profiles. All analysis happens in your browser — the tool does not make outbound HTTP requests to the linked URLs, which makes it fast and private but means you need to separately verify that the flagged URLs actually return errors.

When to replace vs. remove a broken external link

For every broken external link you find, the decision tree is: does a replacement exist at a different URL? If yes, update the link. If no, check whether the Wayback Machine has an archived version — link to `web.archive.org/web/*/url` as an archived citation. If neither exists and the link is only loosely relevant, remove it. Never leave a broken external link in place — it provides negative value to both users and crawlers.

Redirect chains and canonical tag conflicts are two of the most impactful technical SEO errors because they directly affect which page Google decides to index and how much link equity each page accumulates. Neither error produces a visible user-facing problem — both are invisible to visitors — which is why they persist on sites for months or years without being noticed.

Detecting and fixing redirect chains

A redirect chain forms when a URL has a redirect that itself redirects, creating a sequence of hops before reaching the final destination. Google's crawlers follow chains up to a limit, but each additional hop reduces the PageRank passed from the original URL to the destination. A chain of three redirects passes measurably less equity than a single direct 301.

The Redirect Chain Checker analyses your redirect map or server logs and identifies chains longer than a single hop, redirect loops (URL A → B → A), and mixed HTTP/HTTPS sequences. The fix for a chain is always to update the source to redirect directly to the final destination URL, bypassing all intermediate steps.

Canonical tag audit

A canonical tag tells search engines which version of a page is the preferred one for indexing. Common canonical errors include: a page canonicalising to a URL that does not exist, a canonical tag using a relative path rather than an absolute URL, the canonical pointing to an HTTP URL when the site is HTTPS, and a page canonicalising to a different page that itself canonicalises back — a canonical loop. Any of these can cause Google to index the wrong version of a page or to distrust the canonical signal entirely.

The Canonical URL Checker validates canonical tags and URL mappings for all of these conflict patterns. Run it on any page where you suspect indexing problems, on all paginated series pages (which commonly have misconfigured canonicals), and on any page that has been migrated from an old URL.

Your sitemap tells search engines which pages exist and should be crawled. Your robots.txt tells crawlers which paths they are allowed to access. Both files are simple in principle but surprisingly easy to misconfigure — and errors in either can result in important pages being ignored or, worse, actively blocked.

What to check in your sitemap

• Noindex pages included: Any URL in the sitemap that also has a `noindex` robots directive sends a contradictory signal. Remove noindex pages from the sitemap.
• HTTP URLs in an HTTPS site: Every URL in your sitemap should use the HTTPS protocol. An HTTP URL in a sitemap on an HTTPS domain causes the crawler to follow a redirect on every crawl of that URL.
• 4xx and 5xx URLs: Sitemaps should only contain live, accessible pages. A broken URL in the sitemap wastes crawl budget every time Googlebot checks it.
• Missing required elements: The sitemap protocol requires a `<urlset>` wrapper and `<loc>` elements for every URL. Missing these elements renders the sitemap unparseable.
• Exceeding the 50,000 URL limit: A single sitemap file cannot contain more than 50,000 URLs. Use a sitemap index file and multiple chunk files for larger sites.

The Sitemap XML Validator checks all of these structural and protocol compliance issues from a paste of your sitemap file. No URL is fetched — the validation is entirely structural, which makes it safe for sitemaps containing internal URLs you do not want to expose. For sitemaps that are over the size limit, the Sitemap Index Splitter chunks the file into compliant pieces and generates the index file automatically.

robots.txt audit

The robots.txt Disallow directive is a blunt instrument — a single misconfigured rule can block an entire section of your site from being crawled. The most common robots.txt errors are: a `Disallow: /` directive that blocks the entire site (often left over from a staging environment), blocking CSS or JavaScript files that Google needs to render pages, and conflicting Allow/Disallow rules where the more restrictive rule takes precedence unexpectedly.

The Robots.txt Validator checks your robots.txt file for syntax errors, invalid directives, path issues, and structural best practices. Paste the file content — it never requires a live fetch of your domain — and the validator reports every issue with a plain-language description of the potential impact.

A broken link and error audit generates findings across five categories, and not all findings are equal. Trying to fix everything simultaneously is impractical on any site larger than a few dozen pages. A prioritised workflow focuses effort on the errors with the highest SEO and user experience impact first, working down to lower-priority cleanup over subsequent sprints.

Recommended audit order

1. Fix broken internal links on high-traffic pages and navigation menus — these affect both users and PageRank distribution immediately.
2. Fix canonical conflicts on your most important pages — incorrect canonicals cause the wrong URL to be indexed, directly displacing your target page from rankings.
3. Consolidate redirect chains to single-hop 301s — this recovers PageRank that is currently being diluted across the chain.
4. Update or remove broken external links — lower urgency than internal issues but a measurable quality signal.
5. Fix sitemap errors — ensures newly published pages are discovered efficiently.
6. Fix robots.txt errors — critical if any blocking is suspected, otherwise lower urgency than the above.

Building a recurring audit schedule

For sites publishing content regularly, a monthly audit cadence catches accumulation before it compounds. For sites undergoing migrations or redesigns, run an audit immediately before the change to establish a baseline and again immediately after to verify every redirect is in place. Integrating automated checks into your deployment pipeline catches newly introduced broken links before they reach production — most CI/CD frameworks support simple URL validation scripts that can run as a post-deployment step.

Combining link audits with a full SEO health check

A broken link audit is one component of a broader technical SEO review. Once link and redirect issues are resolved, extend the audit to cover meta tags with the Meta Tag Analyzer tools, structured data validity with the Structured Data Validator, and HTML quality with the HTML Validator. These four checks together cover the most common technical SEO issues that prevent well-written content from ranking at its potential.